WORLD POPULATION IN 2050: ASSESSING
PROJECTIONS. DISCUSSION
THE
Cynthia B. Lloyd*
Joel Cohen was asked by the conference organizers to focus a critical
eye on demographic projections and to give special attention to the
outlook for the elderly and for total dependency ratios. He was very
faithful to his assignment. I have only a few points to raise about his
paper before going on to discuss one issue that he did not cover (and was
not asked to cover) but that I think should be addressed as part of a
comprehensive treatment of population prospects for the year 2050. The
issue is the current youth bulge in many of the world’s poorest countries
that have yet to reap the economic benefits of the demographic transition.
First to Cohen’s paper.
1998
VERSUS
2000 UNITED NATIONS PROJECTIONS
Cohen’s paper is based on the 1998 estimates and projections of the
United Nation’s Population Division. The highlights of the 2000 revision
of these projections were released to the public several months ago and
are available on their web site. Indeed, the press took some interest in the
announcement, as this latest revision suggests that there may be 413
million more people living in the world in 2050 than previously projected.
Upward revisions to estimates of recent fertility in Bangladesh, India, and
Nigeria make up 32 percent of this increase. Higher future fertility levels
projected for some sixteen developing countries whose levels have not
yet shown signs of sustained decline explain much of the rest. Obviously
these changes, which primarily affect estimates for developing countries,
somewhat reduce the estimated rise in the median age of the world’s
*Director, Social Science Research, Population Council.
WORLD POPULATION IN 2050: ASSESSING THE PROJECTIONS: DISCUSSION
115
population, from 37.8 to 36.2 by the year 2050. Underlying this revision
are some changes in assumptions as well. The United Nations Population
Division no longer specifies that all high-fertility countries reach replacement fertility by 2050, and its estimates of cross-border migration flows
were based on a more systematic analysis of available data on recent
experience with migration. Let us hope that all the data will be released
soon so that we can all be up-to-date.
THE PROCESS
OF
URBANIZATION
In introducing the topic of urbanization, Cohen talks about the “rush
of people from the countryside to cities.” In some of the background
papers for a current National Academy of Sciences panel on urbanization
in developing countries, a different picture is emerging. Using data from
sixty-nine recent Demographic and Health Surveys (DHS), Montgomery
and Hewett (2001) estimate that no more than 36 percent of recent urban
migrants (those who have arrived in the last five years) have moved
directly from rural areas. Thirty-four percent have moved from other
cities and 30 percent from towns, suggesting a very different process of
migration. These data do not change the conclusion that most future
population growth will occur in urban areas, but they do suggest that
urban growth may be more spread out than previously thought and may
follow a more complex process. Furthermore, their findings suggest that
residents of smaller urban areas will fare more poorly according to a
variety of health indicators than residents of mega-cities. One conclusion
of this study is that the current focus in the literature on the plight of
residents of mega-cities is not warranted by the data.
THE YOUTH BULGE
The title of this conference is “Seismic Shifts: The Economic Impact of
Demographic Change.” I would suggest that the current youth bulge in
many of the poorest countries carries within it the potential for seismic
shifts and future demographic surprises. I see this issue as a gap, not just
in this session but in the conference at large. As this is a topic I have been
working on in recent years, it seems an obvious contribution I can make
to this group. Indeed, it is no accident that the National Research
Council’s Committee on Population has recently initiated a panel on
transitions to adulthood in developing countries. This panel is following
in the footsteps of the panel on demographic projections—a panel on
which several of the distinguished speakers at this conference served: Joel
Cohen, Massimo Livi-Bacci, Ronald Lee, and Michael Teitelbaum.
According to Macunovich (2000), it is only when the size of the youth
cohort begins to rise relative to the size of the adult population, as a result
of declines in child mortality in the early phase of the demographic
116
Cynthia B. Lloyd
transition, that the fertility transition or the transition from large families
to fewer, “better-quality” children begins. However, while she does not
call attention to it in her article, it is important to note that the youth ratio
as she calculates it (the ratio of 15- to 24-year-olds to 25- to 59-year-olds)
appears to have been much lower historically at the time of the fertility
transition in the West than it is in many of today’s poorer countries,
leaving young people in some of the poorer countries today with a much
larger relative disadvantage in terms of cohort size. Thus, in many of the
poorest countries where the fertility transition has not yet begun or is
only beginning, our primary attention should be on the younger end of
the age distribution, not the older end.
The situation of those currently in transition from childhood to
adulthood in the poorer countries—roughly those aged 10 to 24 —is a
neglected topic within this broad discussion of the economic impact of
demographic change. In order to become productive citizens who can not
only support themselves and their families but also become productive
players in the global economy, these young people will need to develop
substantial human and social capital. A lot of the earlier discussion on
what is expected for the year 2050 will depend on what happens to this
next generation of young people as they enter their peak productive and
reproductive years. The adolescent and young adult years are “demographically dense” (Rindfuss 1991). It is during this phase of life that
young people leave school, experience their first job (or their first spell of
unemployment), become socially and legally recognized as adults (the
United Nations Convention on the Rights of the Child defines adulthood
as 18 years old), serve in the military or national service, have their first
sexual experiences, and, particularly for many young women, marry or
have their first birth. It is during this phase of life that young men and
women acquire the human capital and social capital that will shape their
potential to assume economically productive roles in society.
The numbers of young people in the world will peak by the year 2010
in absolute terms (Figure 1). Some 1.5 billion of the 1.8 billion young
people in the world at that time will live in developing countries (United
Nations 1999).1 While in more developed countries young people have
never represented more than 26 percent of the population, in developing
countries this age group will represent 29 percent of the population in
2005. (In developed countries they now represent only 19 percent of the
population.) In the poorest countries, which include most of sub-Saharan
Africa, the relative numbers are even more dramatic, with 33 percent of
the population aged 10 to 24 in 2005 and this percentage projected to
persist for at least another decade (Figure 2). This youth bulge presents
1
I use the 1998 United Nations medium-term projection throughout this discussion.
WORLD POPULATION IN 2050: ASSESSING THE PROJECTIONS: DISCUSSION
117
substantial challenges to the poorer countries in terms of human capital
investment. Furthermore, this age group is so large, in relative terms, that
as it enters the childbearing years it will assure a continued growth of
population even if the next generation of women have only enough
118
Cynthia B. Lloyd
children to replace themselves. Bongaarts (1994) has estimated that
“population momentum,” the component of population growth that is
attributable to a young age distribution due to past high fertility, will
account for about one-half of future population growth in developing
countries.
The only way that the impact of momentum on the population
growth rate can be slowed is through an increase in the age at first birth.
Rising education levels tend to be associated with later ages at first birth.
Knowledge of this link has directed greater attention in recent years to
educational investments as a tool of population policy.
THE IMPORTANCE
OF
HUMAN CAPITAL INVESTMENT
Some of the Asian economies where the fertility transition began in
the 1970s had made substantial and early investments in formal schooling
and were able to reap a demographic bonus (Asian Development Bank
1997) as the youth dependency ratio dropped. But many of today’s poorer
countries face a more daunting challenge in terms of human capital
investments, given the recent rapid growth in school-age populations. In
many sub-Saharan African countries, impressive growth in schooling in
the early post-independence years has been followed by years of stagnation and in many cases recent declines in enrollment and grade attainment.
In 1995, the proportion of 15- to 19-year-olds who had ever attended
school varied from 25 percent in Burkina Faso to 98 percent in Zimbabwe
among the African countries with comparable and accessible household
surveys (we have recent data on twenty-two sub-Saharan African countries in all). Among the countries that perform well on enrollment,
however, is a tremendous range of experience in school retention through
the primary grades. I illustrate this range with five examples from Africa
(Figure 3). You can see that in Mozambique, where 72 percent of 15- to
19-year-olds have attended school, only 12 percent have completed the
five years required for primary school completion. By contrast, in
Tanzania, 85 percent have ever enrolled and 67 percent have completed
the seven years required for primary completion. Both countries are
estimated to have GDP per capita below $500, so differences have to be
explained largely by differences in the share of public resources directed
to primary schooling, not by levels of income.
Several reasons account for the substantial attrition among children
who start school. First, many children start late. Second, many children
are forced to repeat if their attendance record has been poor, if they move,
or if they are unable to pay their school fees (a growing problem in recent
years as many countries have reintroduced school fees in response to
various negative economic shocks), or if the teacher has been absent and
the school shut down—a none too infrequent occurrence. The result is
WORLD POPULATION IN 2050: ASSESSING THE PROJECTIONS: DISCUSSION
119
that few kids progress normally through the school system, starting on
time and moving from grade to grade on an annual basis.
We can illustrate with the case of Kenya (Figure 4). You can see that
while children are meant to start school at the age of 6, children continue
to enter the system until age 10 to 11. These children remain behind grade
for their age throughout their educational careers, while they and many
others are often forced to repeat grades because of economic hardships.
Evidence from a variety of studies shows that children who start late
ultimately complete fewer grades of school. One can imagine how
difficult it must be to teach when students range so widely in age. In
grade four in Kenya, students range in age from 8 to 18 (Kenyan 1998
DHS household survey). It is for these reasons, as well as because of the
shortage of places in secondary school, that the majority of the schoolgoing population in most of these countries is enrolled in primary school.
There is some controversy in the literature about Africa’s economic
performance in the 1990s. While Collier and Gunning (1999) take the
region’s poor performance as a given and try to explain it, Sahn and Stifel
(2000) use alternative data from the DHS on household assets to conclude
that there was some lessening of poverty in Africa during the 1990s.
Trends in educational enrollment and attainment provide another angle
of vision on Africa’s economic performance that might give more weight
to Collier and Gunning’s perspective. It appears that, maybe for the first
time in history, there have been actual declines in educational enrollment
and attainment in many countries of sub-Saharan Africa, particularly for
120
Cynthia B. Lloyd
boys (Lloyd, Kaufman, and Hewett 2000) (Figures 5 and 6). Countries that
made impressive investments in schooling during the early years of
independence (such as Kenya, Tanzania, and Zimbabwe) have been
especially hard-hit. Given growing evidence from diverse settings that
WORLD POPULATION IN 2050: ASSESSING THE PROJECTIONS: DISCUSSION
121
ˆ
enrollments can be responsive to small positive changes in income,2 these
recent trends would strongly suggest that incomes are falling, costs of
schooling are rising, or both. These trends will inevitably accentuate
existing inequalities between rich and poor in the achievement of
universal or basic schooling (Filmer and Pritchett 1999).
The accuracy of population projections to the year 2050 fundamen-
2 See, for example, Schultz (2001) in the case of Progresa in Mexico; Arends-Kuenning
and Amin (2000) and Ravallion and Wodon (1999) in the case of a food for education
program and a secondary school scholarship scheme for girls in Bangladesh; Kremer et al.
(1997) in the case of subsidies for school textbooks and uniforms in Kenya.
122
Cynthia B. Lloyd
tally depends on today’s investments in the human and social capital of
young people. Assumptions about improved mortality, continuing
steady declines in fertility, and modest but persistent levels of migration
are tied to unstated assumptions about a unidirectional process of
development, that is, a steady increase in educational attainment of the
next generation, obviously accompanied by some steady economic
growth. To the extent that some developing countries progress smoothly
toward universal schooling with the support of a growing economy and
others fail to progress or fall behind, global inequalities will become
accentuated, with implications for many aspects of human well-being
that could be manifested in some demographic surprises, possibly
unpleasant ones.
References
Arends-Kuenning, Mary and Sajeda Amin. 2000. “The effects of schooling incentive
programs on household resource allocation in Bangladesh.” Population Council Policy
Research Division Working Paper No. 133.
Asian Development Bank. 1997. Emerging Asia; Changes and Challenges. Manila.
Bongaarts, John. 1994. “Population policy options in the developing world.” Science 263, no.
5148, pp. 771–76.
Collier, Paul and Jan Willem Gunning. 1999. “Explaining African Economic Performance.”
Journal of Economic Literature 37, March, pp. 64 –111.
Filmer, Deon and Lant Pritchett. 1999. “The effect of household wealth on educational
attainment: Evidence from 35 countries.” Population and Development Review 25(1), pp.
85–110.
Kremer, Michael, Sylvie Moulin, David Myatt, and Robert Namunyu. 1997. “Textbooks,
Class Size and Test Scores: Evidence from a Prospective Evaluation in Kenya.” IPR
Working Paper Series, Institute for Policy Reform. Washington, DC.
Lloyd, Cynthia B., Carol E. Kaufman, and Paul Hewett. 2000. “The spread of primary
schooling in sub-Saharan Africa: Implications of fertility change.” Population and
Development Review 26(3), pp. 483–515.
Macunovich, Diane J. 2000. “Relative cohort size: source of a unifying theory of global
fertility transition?” Population and Development Review 26(2), pp. 235– 62.
Montgomery, Mark and Paul Hewett. 2001. Results from the DHS; A Guidebook. Draft.
Ravallion, Martin and Guentin Wodon. 1999. “Does child labor displace schooling?
Evidence on behavioral responses to an enrollment subsidy.” World Bank Policy
Research Working Paper No. 2166. Washington, DC: The World Bank.
Rindfuss, Ronald R. 1991. “The young adult years: diversity, structural change and fertility.”
Demography 28(4), pp. 493–512.
Sahn, David E. and David C. Stifel. 2000. “Poverty comparisons over time and across
countries in Africa.” World Development 28(12), pp. 2123–55.
Schultz, T. Paul. 2001. “School Subsidies for the Poor: Evaluating A Mexican Strategy for
Reducing Poverty.” Paper presented at the Population Association Meetings in
Washington, DC.
United Nations. 1999. World Population Prospects; The 1998 Revision. New York.
United Nations Population Division. 2001. World Population Prospects; The 2000 Revisions,
Highlights. 28 February 2001.